→ view this video on tudou.com (Adobe Flash Plugin required for Safari users)
Welcome to Vita.
With the proliferation of indoor positioning systems, analytics on users' indoor mobility data has emerged as a promising research frontier. However, the database community is missing a synthetic data generator that analysts could use to study algorithms assuming different indoor positioning technologies. Such a generator is needed for two reasons: (1) It can be used to simulate high volumes of indoor mobility data to evaluate location-dependent queries/algorithms. (2) It can provide the additional "ground truth" for the synthetic mobility data generated in (1) in spite of the discrete nature of the simulated data, which is useful for effectiveness studies in analytics.
We demonstrate a generic, user-configurable toolkit for generating different types of indoor mobility data for real-world buildings. Our prototype generates the desired data in a three-layer pipeline: 1) The Infrastructure Layer accepts industry-standard digital building information (DBI) files to generate the host indoor environment, allowing for users to configure the generation of a variety of positioning devices, such as Wi-Fi, Bluetooth, RFID etc. 2) The Moving Object Layer offers the functionality of defining a set of objects or trajectories, with configurable indoor moving patterns, distribution models as well as sampling frequencies. 3) The Positioning Layer generates synthetic signal strength measurements known as raw RSSI measurements according to the positioning device data and trajectory data generated at relevant layers. It also generates different types of indoor positioning data through the customization of all typical indoor positioning methods on the raw RSSI data.
Vita is characterized by several desirable features:
- Vita accepts industry-standard DBI files and uses real-world (multi-floor) buildings with semantic attributes as the host environment for data generation.
- Vita offers a set of indoor moving patterns and distribution models for users, making its output---indoor moving objects and their trajectory data---highly configurable.
- Vita allows users to tune the sampling frequency of generating the trajectories such that the valuable "ground truth" movements are preserved to an arbitrarily detailed degree.
- Vita can generate indoor positioning data according to the most typical indoor positioning methods.
Authors and Contributors
- Huan Li, Zhejiang University, China
- Hua Lu, Aalborg University, Denmark
- Xin Chen, Zhejiang University, China
- Gang Chen, Zhejiang University, China
- Ke Chen, Zhejiang University, China
- Lidan Shou, Zhejiang University, China
Support or Contact
Email Address: lihuancs@zju.edu.cn
Vita is designed and implemented in an extensible way for easy integration of more advanced features in the future. The entire prototype are now being open source for free extensions and customizations. You are welcome to contribute to Vita, to make it better :)
If you feel our project is useful for your research or development, please cite our paper:
Li, H., Lu, H., Chen, X., Chen, G., Chen, K., & Shou, L. (2016). Vita: A versatile toolkit for generating indoor mobility data for real-world buildings. Proceedings of the VLDB Endowment, 9(13), 1453-1456.